Intelligent approaches for the synthesis of petrophysical logs

M. Reza Rezaee, Ali Kadkhodaie-Ilkhchi, Pooya Mohammad Alizadeh

Research output: Contribution to journalArticlepeer-review

54 Citations (Scopus)

Abstract

Log data are of prime importance in acquiring petrophysical data from hydrocarbon reservoirs. Reliable log analysis in a hydrocarbon reservoir requires a complete set of logs. For many reasons, such as incomplete logging in old wells, destruction of logs due to inappropriate data storage and measurement errors due to problems with logging apparatus or hole conditions, log suites are either incomplete or unreliable. In this study, fuzzy logic and artificial neural networks were used as intelligent tools to synthesize petrophysical logs including neutron, density, sonic and deep resistivity. The petrophysical data from two wells were used for constructing intelligent models in the Fahlian limestone reservoir, Southern Iran. A third well from the field was used to evaluate the reliability of the models. The results showed that fuzzy logic and artificial neural networks were successful in synthesizing wireline logs. The combination of the results obtained from fuzzy logic and neural networks in a simple averaging committee machine (CM) showed a significant improvement in the accuracy of the estimations. This committee machine performed better than fuzzy logic or the neural network model in the problem of estimating petrophysical properties from well logs.

Original languageEnglish
Pages (from-to)12-26
Number of pages15
JournalJournal of Geophysics and Engineering
Volume5
Issue number1
DOIs
Publication statusPublished - 1 Mar 2008
Externally publishedYes

Keywords

  • Artificial neural networks
  • Committee machine
  • Fuzzy logic
  • Petrophysical logs
  • Synthesizing

Fingerprint

Dive into the research topics of 'Intelligent approaches for the synthesis of petrophysical logs'. Together they form a unique fingerprint.

Cite this